Removing the brain part, as the epilepsy source attack, is a surgery solution for those patients who have drug resistant. So, the epilepsy localization area is an essential step before surgery. The Electroencephalogram (EEG) signals of these areas are different and called as focal (F) where the EEG signals of other normal areas are known as non-focal (NF). Visual inspection of multi channels for detecting the F EEG signal is time consuming and along with human error. In this paper, we propose a new method based on ensemble empirical mode decomposition (EEMD) in order to distinguish the F and NF signals. For this purpose, EEG signal is decomposed by EEMD and the corresponding intrinsic mode functions (IMFs) are obtained. Then various nonlinear features including log energy (LE) entropy, Stein's unbiased risk estimate (SURE) entropy, information potential (IP) and centered correntropy (CC), are extracted. At the end, the input signal is classified as either F or NF by using support vector machine (SVM). Using nonlinear features, we achieved 89% accuracy in classification with tenfold cross validation strategy.
Synthetic aperture radar (SAR) images are inherently affected by multiplicative speckle noise. Two approaches based on modeling the nonsubsampled Shearlet transform (NSST) coefficients are presented. Two-sided generalized Gamma distribution and normal inverse Gaussian probability density function have been used to model the statistics of NSST coefficients. Bayesian maximum a posteriori estimator is applied to the corrupted NSST coefficients in order to estimate the noise-free NSST coefficients. Finally, experimental results, according to objective and subjective criteria, carried out on both artificially speckled images and the true SAR images, demonstrate that the proposed methods outperform other state of art references via two points of view, speckle noise reduction and image quality preservation.
We propose two methods for speckle suppression of synthetic aperture radar (SAR) images. The first method is based on Bayesian shrinkage and is a thresholding technique. The main problem of applying Bayesian shrinkage in a transformed domain, such as contourlet transform (CT), is finding the optimum threshold value. According to our experimental results, contourlet coefficients are affected by noise differently. It means that some contourlet coefficients belong to the specific sub-bands that are more robust against noise. We use this newfound property to determine the optimum threshold value and to develop our proposed method, which is named the weighted Bayesian shrinkage in contourlet domain. The second method, named the NSCT-GΓD, is a model-based approach using a two-sided generalized Gamma distribution (GΓD) to model the statistics of nonsubsampled contourlet transform (NSCT) coefficients. We use the Bayesian maximum a posteriori estimator to find NSCT despeckled coefficients. Experimental results carried out on both artificially speckled images and the true SAR images show that our two proposed methods outperform other approaches via two point of views, speckle noise reduction and image quality preservation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.